There is something to this, Michaelg. Seems to filter out much of the rubbish, and also false favourites. Are you getting better results from Saturday metrop meetings? I am seeing a slight Place profit on main Saturday meetings, if bet 'best tote'.

Hi, Jackact.

Unfortunately I have only last Saturday, which using the current rules the result is:

13 selections for 11 placegetters giving a POT of 53.8% with NSW TAB divvies.

In comparison, since Sunday there have been 15 selections for 11 placegetters for a POT of 10.0%

I could manually look at Saturday results for the past few months to get some sort of idea, but that would take a long time. I presume you have some results for Saturdays?

Michael, i will run it thru later today & post back

I don't have a data base, Michaelg, but I did choose five Saturdays over the past few months: 14 March, 18 April, 25 April, 2 May and 9 May, and did them manually. Overall POT for a place (Sportsbet best price) was just over 4 per cent. I did not use your filter of place strike rate of 30% +, so may have been better.
4 per cent is not great, but may indicate this is a good general filter for 100 pointers. Other filters may be able to lift the POT.

total races reviewed - 3973
selections - 772

wins - 195
s/r - 25.3%
collect $632
bet $772
net -$140
lot -18.1%

places - 417
s/r - 55.8%
collect $654
bet $772
net -$118
lot -15.3%

Thanks, Stugots.

To recap the rules:

1) 8 to 12 starters.
2) Outright top selection in Unitab ratings.
3) Last two placed-positions to be between 3 and 6 points.
4) Unbroken by a spell.
5) Minimum Place strike rate of 30%.
6) No Maidens, Hurdles and Highweights (races with Hurdle jockeys)

Jackact, there have been several selections whose Place strike rate has been under 30%. They have had a poor strike rate and were not in profit so I've filtered them out of the method.

Today's selections

Morphetville R5 no.5 - Silver Shamrock
Kembla R6 no.8 - Tully Endeavour
Belmont R1 no.6 - New Image
Belmont R3 no.5 - Beyond Repute
Belmont R4 no.11 - Strait Laced
Sun Coast R1 no.2 - The Lording Dragon

& just out of interest the heaviest losing day for the win was Saturdays - lot of 37%

I have spent a bit of time looking for a couple of filters but nothing stands out amongst the usual suspects - weight carried or +/-, distance, age & sex, win%, track & distance records, days last start etc

good luck with it tho, hope you can pickup on something

Thanks heaps, Stugots. Looks like I used too small a sample of Saturdays. Yesterday was not flash, either.
I still think Michaelg is onto something here. His main filters (the first four in his last post) put Top Raters into two categories I reckon: Good Value (his selections) and Bad Value (the Top Raters he filters out).
If we can't turn a consistent profit on the Good Values, maybe we can on the Bad Values. For example:
-only Metrop meetings
-only races where only one Top Rater
-only where the Top Rater takes a sizeable percentage of the pool, e.g. win price of $5.00 and below
-Lay on Betfair, or bet around the Top Rater.

I'll do some more work and let you guys know if I find anything that works consistently.

Thanks, Stugots.

The loss for Saturdays of 37% is almost double the loss if every 100 pointer was layed - it seems extraordinarily high.

Interestingly, my figures show that if all the selections since last Sat (8 days) had been layed then it would be in profit.

Today's selections are:

Sun Coast R1 no.4 - unplaced
Sun coast R7 no.8
Cranbourne R6 no.4
Wellington R3 no.2 - ran second
W'bool R6 no.7

Might be a Lay system.

None of today's five selections won.

If layed they would have produced a profit of 4.75 units.

I'll now record it as a Lay method.